The liver possesses profound regenerative potential. Following acute injury, even if a large part of this organ is destroyed, the liver is capable of restoring its complete mass and original architecture in a comparatively short time. In contrast, chronic progressive injury to the liver will induce repetitive tissue damage and result in impaired regenerative capacity marked by sustaining inflammation and extracellular matrix remodeling, eventually culminating in fibrosis and cirrhosis (1). Regardless of the underlying cause, liver fibrosis is viewed as a wound healing response initiated by inflammation responding for different iterative parenchymal damages (2). This injury-induced inflammation has been proved as a double-edged sword for liver regeneration and repair: moderate and resolved inflammatory responses are beneficial to the liver as they exert consistent hepato-protective effects, contribute to tissue repair, and promote the recovery of tissue homeostasis, whereas excessive and persistent inflammatory responses not only aggravate liver injury as triggered by a wide array of hepatotoxic insults but also promote fibrosis and even hepatic carcinogenesis (3). Therefore, the injury-induced inflammation has an important impact on regulating liver regeneration and fibrosis. The mechanisms that govern fibrogenic liver repair are complex and interactions between various types of innate and adaptive immune cells with stromal cells mediated by cytokines, growth factors, or modified matricellular proteins are implicated (4). Understanding the immunological mechanisms of liver fibrosis provides the foundation for designing new therapies that do not depend on the underlying cause.

Several signal transduction pathways, which direct cell fate decisions including cell proliferation, differentiation, and migration in the phase of embryogenesis, are reactivated in the case of liver injury and are verified to regulate liver regeneration and repair. One such signal transduction pathway is hedgehog (Hh) signaling pathway. Hh pathway, initially verified in the Drosophila fruit fly, is a highly conserved signaling pathway, which is considered to play an important regulatory role in embryonic development and adult tissue remodeling of multiple tissues and organs, including the liver (5). In particular, evidence from studies in adult rodents and humans indicates that the injury-related activation of Hh signaling coordinates several fundamental facets of fibrogenic liver repair, such as the amplification of liver progenitors, the aggregation of hepatic myofibroblasts, repair-related inflammatory responses, and vessel remodeling (6). Noticeably, the regulation of repair-related inflammatory responses in chronic liver injury by Hh signaling pathway extends current understanding of this signaling pathway’s immune-modulatory actions, since the immune regulation of Hh signaling pathway initially reflects that Hh ligands have been identified as key modulators of many phases of T-cell development in the thymus (7). Growing evidence from a number of different studies in vivo and in vitro has indicated that immune-mediated events involved in liver injury and fibrogenesis are regulated by Hh signaling pathway. Hence, in this review, we emphasize that Hh, as a non-immune factor existing on the microenvironment of tissue repair and regeneration, has been shown to modulate immunological aspects in the fibrotic process of chronic liver diseases. More importantly, these data contribute to characterize how microenvironmental cues influence immune cells to modulate liver tissue remodeling. Furthermore, we also discuss that Hh pathway inhibitors are good candidates for anti-fibrotic therapeutic agents because of their immune-regulation actions for fibrogenic liver repair.

Hedgehog signaling pathway is a highly complex signaling pathway and mainly involves canonical signaling pathway and non-canonical signaling pathway. There are three types of Hh ligands, Sonic Hedgehog (Shh), Indian Hedgehog (Ihh), or Desert Hedgehog (Dhh) in Hh signaling pathway (8). In canonical signaling pathway, the binding of Hh ligands (Shh, or Ihh, or Dhh) to the transmembrane receptor patched (Ptc) triggers Ptc to relieve its tonic inhibition of Smoothened (Smo) (9). Released Smo then aggregates in the primary cilium, which contributes to the nuclear localization of Gli transcription factors (9). There are three glioma-associated oncogene homolog (Gli) transcription factors (Gli1, Gli2, and Gli3), all of which are zinc finger proteins comprising a DNA-binding domain and five tandem C₂H₂ zinc finger motifs (8). Meanwhile, there is an activator domain (GliA) at the C-terminus of all three Gli transcription factors and is a repressor domain (GliR) at the N-terminus of Gli2 and Gli3 (8). In mammals, the function of three Gli transcription factors is partly overlapping but also contrary. Gli2 is characterized as the main activator of Hh signaling, whereas Gli3 is accountable for the repression function in response to Hh signaling (9). In the nucleus, Gli2 or Gli3 binds to DNA and then modulates the transcription of numerous Hh target genes. As for Gli1, it seems to be a signaling amplifier of the Gli2-mediated transcriptional response (9). Suppressor of fused (SUFU) is an important negative modulator of Hh signaling. Lack of Hh ligands, the binding of SUFU to Gli transcription factors facilitates the ubiquitination and degradation of Gli transcription factors, and correspondingly inhibits the activation of Hh signaling pathway. In addition, there is a vertebrate-specific feedback antagonist of Hh signaling, Hh-interacting protein (Hhip), which competes with Ptc for binding to Hh ligands and then attenuates Hh signaling through a negative regulatory feedback loop (10). Besides the aforementioned canonical signaling pathway, numerous studies suggest that there are non-canonical signaling pathways. For example, Ptc is shown to modulate cell cycle through cyclin B1, which do not require Smo and Gli transcription factors (11, 12). Moreover, the downstream effect of Smo also is demonstrated to be mediated by the activation of small GTPases independent of Gli transcription factors (13).

During fibrogenic liver repair caused by various types of chronic injury, emerging evidence has demonstrated a critical role of canonical Hh signaling, which is supported that conditional deletion of Smo in α-SMA⁺ myofibroblasts inhibited liver fibrosis (14). Moreover, multiple injury-associated factors participating in liver fibrogenesis have been shown to directly activate Gli transcription factors, which include platelet-derived growth factor (PDGF) (15), transforming growth factor β1 (TGF-β1) (16), epidermal growth factor (EGF) (17), and insulin-like growth factors (18). Conversely, Hh transcription factors have been demonstrated to regulate the activities of target genes in other signaling pathways. For example, Gli could regulate transcription of a TGF-β/Smad signaling pathway target gene Snail (19) and also impact the expression of Wnt signaling pathway modulators, such as Wnt5a (an activator of Wnt signaling pathway) and soluble frizzled receptor-1 (sFRP1, an inhibitor of Wnt signaling pathway) (20). Therefore, Hh signaling pathway, as one of key regeneration signaling (14) and fibrogenic signaling pathways (21), is part of a complex signaling network that involves other essential signaling pathways controlling cell fate acquisition, such as TGF-β/Smad and Wnt signaling pathways. Within this signaling network, Hh signaling interacts with these signaling pathways, which orchestrates global changes in the phenotypes and function of Hh-responsive cells and regulates the injured liver regeneration and fibrogenesis (22).

In healthy adult liver, there is lack of detectable activity of Hh pathway because of low Hh ligand expression and relatively high Hhip expression (22). During liver injury, injury stimulates liver epithelial cells, including mature hepatocytes (23, 24) and cholangiocytes (25), to produce and release Hh ligands, and simultaneously represses sinusoidal lining cell expression of Hhip (17, 26, 27). Then, Hh ligands activate the Hh signaling pathway in surrounding Hh-responsive cells, including quiescent hepatic stellate cells (Q-HSCs), liver progenitors, liver sinusoidal endothelial cells (LSECs), and immune cells, through ligand–receptor interactions. Meanwhile, most of these Hh-responsive cells are also Hh-producing cells, which are capable of producing Hh ligands in response to injury-induced Hh ligands or other injury-associated factors [e.g., PDGF-BB (15), TGF-β1 (16), and EGF (17)] in turn. This triggers a cascade causing excessive activation of Hh signaling pathway. Because of the extensive enrichment of Hh ligands, the proximity of Hh-producing and Hh-responsive cells in anatomical structure, with the addition of the complex autocrine and paracrine Hh signaling loops in the remodeling and reconstruction process of injured liver, Anna Mae Diehl’s group (28) and Grzelak and colleagues (29) put forward the concept of the intrahepatic signaling niches of Hh. In this injury-related intrahepatic signaling niches of Hh, Hh signaling as a mediator has an impact on the interactions between Hh-producing cells and Hh-responsive cells, as well as their biological functions in liver regeneration and repair (30).

Existing data suggest that determining whether Hh signaling induces adaptive responses or maladaptive responses to liver injury depends on the equilibrium between its regulatory effect on the expansion of progenitor cells and its impact on promoting liver inflammation and fibrosis (22). Hh signaling activated by injury is critical for the regeneration of liver tissue, which is verified that treatment with Smo antagonist interfered with the accumulation of liver progenitors, suppressed the proliferation of hepatocytes and cholangiocytes, blocked the regeneration of liver tissues, and dramatically reduced the survival of normal mice after 70% partial hepatectomy (31). However, due to persistent liver injury, the sustained expression of Hh signaling perpetuates the expansion of cell types, which are accountable for the fibrotic progression of chronic liver diseases, including myofibroblasts, activated sinusoidal endothelial cells, immature liver epithelial cells, and certain types of immune cells (15, 16). Especially, myofibroblasts are viewed as the major source of fibrillar collagen forming the fibrous scar in fibrotic liver tissues, and the myofibroblastic transformation of resident Q-HSCs is commonly considered the primary origin of myofibroblasts in most cases of chronic liver injury (32). Study showed that Hh ligands can activate hepatic stellate cells (HSCs) and induce their transdifferentiation from a quiescent phenotype into a myofibroblastic phenotype responsible for matrix deposition (22). Moreover, the activation of Hh pathway can inhibit apoptotic signals, enhance the viability and proliferative capacity of myofibroblastic HSCs (MF-HSCs), and stimulate additional production of endogenous Hh ligands in an autocrine or paracrine manner, which drives a positive feedback loop to amplify Hh signaling (17). Meanwhile, accumulating Hh ligands also induce the generation of hepatic myofibroblasts by non-conventional means. This mainly reflects on the following two aspects: first, activating Hh signaling can induce the secretion of monocyte chemoattractant protein 1 (MCP-1) and IL-13, which can chemoattract circulating monocytes/fibrocytes derived from bone marrow into injury liver and promote the transformation of monocytes into fibrocytes, respectively (33, 34). Second, growing evidence shows that Hh signaling can induce epithelial–mesenchymal transition (EMT) in certain liver epithelial cells, such as immature ductular cells (35), as part of repair following chronic injury in adult liver. Therefore, excessive Hh signaling plays a major role in hepatic accumulation of myofibroblasts in liver fibrogenesis, and pharmacologic inhibition of Hh signaling has been shown to alleviate liver fibrosis by abolishing the accumulation of myofibroblasts and resolving collagen matrix in the model of chronic cholestatic liver injury (36).

As mentioned above, liver fibrosis is commonly preceded by chronic inflammation (37). Excessive Hh signaling activated by chronic injury in the progression of liver fibrosis has been shown to modulate hepatic accumulation of inflammatory cells. Hh ligands are produced by a variety of cells including dying hepatocytes, cholangiocytes, (myo-)fibroblasts, and immune cells, and then stimulate immature cholangiocytes to secrete chemokines through autocrine or paracrine modes of action, which can further recruit, attract and retain immune cells (e.g., macrophages, monocytes, neutrophils, and lymphocytes) into the injured liver tissues (33). This might explain why fibrogenesis takes place under the circumstance of chronic inflammation, and this inflammation throughout the whole process of fibrogenic liver repair is consequently described as the “repair (fibrosis)-associated inflammatory response” (6, 37).

Moreover, immune cells participating in this repair-associated inflammation have been considered as pivotal players in the fibrogenic response and can exert their ability of either inducing injury or promoting repair (38). Most of these immune cell populations, consisting of classic immune cell populations [e.g., macrophages/Kupffer cells, natural killer T (NKT) cells, lymphocytes, neutrophils, and monocytes] and non-classical immune cell populations (i.e., non-hematopoietic cell populations, including LSECs, HSCs, hepatocytes, and cholangiocytes), are either Hh-producing cells or Hh-responsive cells (30) (Table 1). Increasing study further demonstrated that besides promoting the accumulation of inflammatory cells within the injured liver tissue, excessive Hh signaling in the fibrotic microenvironment also could modulate the phenotype of immune cells as well as their function. Hence, we will emphasize the impacts of injury-activated Hh signaling on liver fibrogenesis through modulating repair-related inflammation, and focus on the regulatory action of aberrant Hh signaling on repair-related inflammatory responses mediated by the aforementioned classical and non-classical immune cell populations in the progression of liver fibrosis (Figure 1).

Recently published literature describes Hh pathway inhibitors as the following three categories: inhibitors targeting Smo, including nature compound (cyclopamine) and synthetic small molecules (vismodegib, sonidegib, IPI-926, BMS-833923, PF-04449913, SANT1–4, Cur61414, MRT-10); strategies to target the upstream N-Shh ligand, including Robotniknin and anti-Shh 5E1; and inhibitors targeting the downstream transcription factors Gli1 and Gli2, including Darinaparsin, GANT-61/58, HPI-1, and Forskolin (90). Various in vivo and in vitro studies demonstrated that some of these Hh pathway inhibitors, such as vismodegib (91), sonidegib (92), Forskolin (93), GANT-61 (94), GANT-58 (95), and cyclopamine (96), are capable of inhibiting or reversing the progression of liver fibrosis. Moreover, modulating repair-related inflammation response during the liver injury and repair is one of the important mechanisms of action for anti-fibrotic effect of these Hh pathway inhibitors. Study based on various rodent models of NASH induced by diet suggested that specific Smo inhibitors (vismodegib or sonidegib) suppressed the activity of Hh signaling and accordingly alleviated the degree of fibrosis, mainly by attenuating macrophage-associated and NKT-related liver inflammation (39, 44, 97, 98). Consistent with the above data in NASH, the anti-fibrotic effect of vismodegib was evidenced by significant amelioration of hepatic schistosomal fibrosis in a murine experimental model, which was associated with the significant reduction in granuloma diameter and amelioration of the liver tissue affection seen in mild inflammatory cellular infiltrate and portal tract affection (91). Moreover, in vitro studies demonstrated that two different pharmacologic antagonists of Smo (cyclopamine or vismodegib) also alleviated SEA-induced liver inflammation response by inhibiting the alternative activation of macrophages and consequently contributed to limiting the progression of schistosomiasis fibrosis (42). In addition, in a rat model of CCl₄-induced liver fibrosis, another Hh pathway inhibitor Forskolin can inhibit the Hh signaling pathway through cAMP-mediated activation of PKA, the latter of which, in turn, caused further attenuation of inflammation, repression of HSC activation, and, consequently, decreased liver fibrogenesis (93).

More importantly, Hh signaling has been involved in both liver fibrosis and HCC. It has been proposed that the excessive Hh signaling of HSCs and inflammatory cells, as well as progenitors in injured liver, contributes to the occurrence and development of HCC (23, 99). Recent study based on a Shh-transgenic mouse model further revealed that the aberrant activation of Hh signaling in the stromal cells and immune cells contributes to creating a microenvironment favoring the malignant transformation of hepatocytes during liver fibrogenesis (100). This seems to, to some extent, explain why chronic liver injury in accompany with fibrogenesis is a main risk factor for the occurrence of HCC. Moreover, treatment with vismodegib in the Mdr2⁻/⁻ mouse model of progressive liver fibrosis and spontaneous hepatocarcinogenesis significantly relieved the degree of fibrosis, induced the involution of intrahepatic HCC, and reduced the number of metastatic HCC through inhibiting hepatic Hh signaling activity and decreasing liver myofibroblasts and progenitors without increasing mortality (101). Therefore, Hh-targeted pharmacological interventions might serve as a potential therapeutic strategy not only for liver fibrosis but also for fibrosis-associated HCC.

Since anti-fibrotic strategies based on manipulating the immune system are beginning to emerge in the clinical field (38), it is timely and necessary to clarify how immune responses may be subtly modified in liver tissue remodeling. The conception that Hh, as a morphogen existing on liver injury microenvironment, could manipulate the immunologic aspects of liver fibrotic repair and affect the pathologic inflammatory milieu within liver fibrogenic tissues, is now more and more strongly supported by fantastic amounts of data. This largely contributes to enlarging our understanding about the pathogenesis of liver fibrosis and promotes the development of Hh-centered therapeutic approaches. Although Hh signaling pathway is shown to hold central roles for promotion of liver inflammation and fibrosis, as well as in the transition from chronic injury to hepatocarcinogenesis, the precisely molecular determinants in mediating these roles are to a large extent unknown. Meanwhile, more details about how Hh signaling is modulated by and interacts with other morphogen signaling (e.g., Wnt and Notch) that also regulate adult liver repair should be further investigated. In addition, more studies should be required in order to evaluate the efficacy of Hh pathway inhibitors as anti-fibrotic agents in various experimental liver fibrosis. If successful, it might pave the way to establishing Hh-centered therapies for liver fibrosis and fibrosis-associated HCC, especially both the effectiveness of these inhibitors in solid organ tumors (e.g., advanced basal cell carcinoma) (102) and their human safety profile (103, 104) have been verified in clinical trials. Taken together, based on the aforementioned data, we conclude that detailed knowledge of modulating the repair-associated inflammation by the injury-related activation of Hh signaling pathway might deepen our understanding of the relationship between inflammation and fibrosis underlying liver injury and repair and have potential clinical application in the therapy of liver fibrosis.

XS and HL researched the topic. XS and YP wrote and edited the manuscript. All authors have read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.